Abstract
The Spatial-Stochastic Neural Networks Model (SSNNM) is used to estimate long-term streamflow in parallel reservoir groups. SSNNM employs two kinds of backpropagation algorithms, separately based on LMBP and BFGS-QNBP. SSNNM has three layers in the structure-input, hidden, and output. The network configuration consists of 8-8-2 nodes in each one. Nodes in the input layer are composed of streamflow, precipitation, evaporation, and temperature with monthly average values collected from the Andong and Imha reservoirs. But some temporal differences apparently exist in their time series. For the SSNNM training procedure, the training sets in the input layer are generated by the PARMA (1,1) stochastic model and cover insufficient time series. Generated data series are used to train SSNNM and the model parameters, optimal connection weights and biases estimated during training procedure. They evaluate model validation using observed data sets. In this study, by comparing statistical analysis and hydrographs in the model validation, we find that the new approaches give outstanding results. SSNNM will help manage and control water distribution and provide basic data to develop a long-term coupled operation system in parallel reservoir groups of the upper Nakdong River, South Korea.
Similar content being viewed by others
References
Battiti, R. (1989). “Accelerated back-propagation learning: two optimization methods.”Complex System, Vol. 3, pp. 331–316.
Battiti, R. and Masulli, F. (1990). “BFGS optimization for faster and automated supervised learning.”Proc. of Int. Neural Network Conf. (INNC 90), Paris, France, pp. 757–760.
Coulibaly, P., Anctil, F., and Bobèe, B. (2000a). “Daily reservoir inflow forecasting using artificial neural networks with stopped training approach.J. Hydro., Vol. 230, pp. 244–257.
Coulibaly, P., Anctil, F. and Bobèe, B. (2000b). “Neural network-based long-term hydropower forecasting system.”J. Comp. Aided Civ. And Infrastruct. Engrg., Vol. 15, No. 5, pp. 355–364.
Demuth, H. and Beale, M. (2000).Neural network toolbox: for use with the MATLAB user's guide. The Math Works, Inc.
Gallant, S.I. (1993).Neural network learning and expert systems. MIT Press, Cambridge, MA.
Gill, P.E., Murray, W., and Wright, M.H. (1981).Practical Optimization, Academic Press, NY.
Hagen, M.T. and Menhaj, M. (1994) “Training feedforward networks with the Marquardt algorithm.”IEEE Transactions on Neural Networks, Vol. 5, No. 6, pp. 25–32.
Haykin, S. (1994).Neural networks: A comprehensive foundation. Macmillan College Pub. Comp, Inc., MA.
Jain, S.K., Das, D. and Srivastava, D.K. (1999). “Application of ANN for reservoir inflow prediction and operation.”J. Water Resour. Plng. and Mgmt., ASCE Vol. 125, No. 5, pp. 263–271.
Kim, S. (2000a). “A study on the forecasting of daily streamflow using the multiplayer neural networks model.”J of Korea Water Resour. Assoc., Vol. 33, No. 5, pp. 537–550. [In Korean].
Kim, S. (2000b). “The application of the neural networks method for forecasting the flood discharge in the river basin.”J. of Korea Society of Civil Engineers, Vol. 20, No. 6-B, pp. 801–811 [In Korean].
Kim, S. and Cho, J.S. (2002b). Determination of monthly mean inflow using a spatial-stochastic neural networks model in the multivariate reservoir groups.2002 Proc., Korean Society of Civil Engineers, KSCE, Busan, South Korea, pp. 155–158. [In Koream]
Kim, S., Cho, J.S., and Jung J.Y. (2002a). “Steamflow estimation using the stochastic neural networks model in the multivariate reservoir.”2002 Proc., Korea Water Resour. Assoc., KWRA, Incheon, South Korea, pp 93–103. [In Korean]
Kim, S., Cho, J.S., and Park, J.K. (2003). “Hydrological analysis using neural networks in the parallel reservoir groups, South Korea.”Proc., World Water & Environmental Resources Congress 2003, ASCE/EWRI, Philadelphia, PA.
Kim, S. and Lee, S. (2000). “Forecasting the flood stage using neural networks in the Nakdong River, South Korea.”Proc., Watershed Management & Operations Management 2000, ASCE/EWRI, Fort Collins, Co.
Kim, S., Lee, S., and Cho, J.S. (2001). “Hydrological forecasting based on hybrid neural networks in a small watershed.”J. of Korean Water Resour. Assoc., Vol. 34, No. 4, pp. 303–316. [In Korean]
KMA (1980–1990). South Korea: The Book of Yearly Meterological Data.
MOCT, KOWACO (1991). South Korea: Andong and Imha Reservoir Coupled Operation Caused by Conveying Water via the Conduit of the Yeongcheon Reservoir, Final Report.
MOCT, KOWACO (1999). South Korea: Survey project for the water supply capacity of the multivariate reservoir. The Nakdong and Keum River, Final Report.
Nguyen, D.H. and Widrow, B. (1990). “Neural network for self-learning control systems.”IEEE Control Systems Magazine, pp. 18–23.
Salas, J.D. (1998). SAMS:Stochastic Analysis, Modeling, and Simulation User Manual. Colorado State University, Fort Collins, CO.
Salas, J.D., Delleur, J.R., Yevjevich, V., and Lane, W.L. (1980).Applied Modeling of the Hydrologic Time Series, Water Resources Publications. Littleton, CO.
Shanno, D.F. (1978). “Conjugate gradient methods with inexact searches.”Mathematics of Operations Research, Vol. 3, No. 3, pp. 244–256.
Shin, H.S. and Park, M.J. (1999). “Spatial-temporal drought analysis of South Korea based on neural networks.”J. of Koream Water Resour. Assoc., Vol. 32, No. 1, pp. 15–29. [In Korean]
Thirumalaiah, K. and Deo, M.C. (1998). “River stage forecasting using artificial neural networks.”J. of Hydro. Eng., ASCE, Vol. 3, No. 1, pp. 26–32.
Thirumalaiah, K. and Deo, M.C. (2000). “Hydrological forecasting using neural networks.”J. of Hydro. Eng., ASCE, Vol. 5, No. 2, pp. 180–189.
Tokar, A.S. and Johnson, P.A. (1999). “Rainfall-runoff modeling using artificial neural networks,”J. of Hydro. Eng., ASCE Vol. 4, No. 3, pp. 232–239.
Yapo, P.O., Gupta, V.H., and Sorooshian S. (1996). “Automatic calibration of conceptual rainfall-runoff model: sensitivity to calibration.”J. Hydro., Vol. 181, pp. 23–48.
Zealand, C.M., Burn, D.H. and Simonovic, S.P. (1999). “Short term streamflow forecasting using artificial neural networks.”J. Hydro., Vol. 214, pp. 32–48.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Kim, S. Neural networks model and embedded stochastic processes for hydrological analysis in South Korea. KSCE J Civ Eng 8, 141–148 (2004). https://doi.org/10.1007/BF02829090
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF02829090